DEVONIC FISHES OF THE NEW YORK FORMATIONS 



49 



in surface views or in sections, of vertebral centra or arches, and the pres- 

 ervation of the specimens is so perfect that there is every reason to believe 

 such structures, even if formed of cartilage only, were absent. Neither have 

 we found any indication of a notochord, although one may infer from the 

 outline of the trunk that a notochord was present. It was probably sur- 

 rounded by a membranous sheath of no more consistency, if as much, than 

 that in Amphioxus. 



We may now return, finally, to a consideration of one or two doubtful 

 matters. Neither on the dorsal nor ventral surface of the head region are 

 there any certain indications of olfactory openings. It has been suggested 

 by Smith Woodward, however, that the rounded notch occurring at the 

 external angles of the maxillary plates in Pterichthys and Bothriolepis may 

 indicate the opening of a nasal sac, and it is indeed difficult to conceive 

 what other function it could have subserved. Professor Patten offers no 

 explanation of the notches in question, but advances the novel idea that 

 the olfactory and orbital openings were confluent. The pineal plate and 

 the T-shaped one in front of it, alongside of which the olfactory nerves are 

 supposed to have passed according to Patten's conjecture, were capable of 

 being moved forwards and backwards, coincidently with the sclerotics, the 

 eyes and olfactory pits being opened by one set of movements, and closed 

 by a reverse set. An awkward defect of such a contrivance, if it ever 

 existed, is that the creature could not smell with its eyes shut. One might 

 esteem it a doubtful advantage, also, that the same opening in the head 

 shield should have served for both sight and smell. Possibly it is on this 

 account that the ingenious author whom we have just quoted imagines that 

 the eyes " were placed on short stalks attached to the margin of the orbits 

 by flexible membranes. The lateral end of each stalk was convex, covered 

 with a smooth shell, and could evidently be raised above the orbit or 

 lowered into it." ' Comment on these extremely fanciful hypotheses is 

 unnecessary. 



Formation and locality. Upper Devonic ; Scaumenac bay, province of 

 Quebec, Can. 



'Structure of the Ostracoderms. Science, n. s. 1903. 17: 488. 



